Getting distinct values with `groupby` only doesn't work

[Pierlou]

If we want to retrieve the unique values of a column, the best current syntax would be ?<col_name>__groupby, but it doesn't work as expected: without any aggregation operator in the query (like sum or mean), the returned data only contains the value of the col_name column for each row, as they appear in the table.
This can be explained as the Postgrest query is then select=<col_name>, which only selects the values, without performing any additional operation, including a groupby. A current way to achieve a real groupby is to add an aggregation operation to the query.
Example:
https://tabular-api.data.gouv.fr/api/resources/75a43900-a916-4105-adcf-fe88f88194a8/data/?inseecommuneprinc__isnotnull&inseecommuneprinc__groupby
yields

{
  "data": [
    {
      "inseecommuneprinc": "01007"
    },
    {
      "inseecommuneprinc": "01007"
    },
    {
      "inseecommuneprinc": "01007"
    },
    {
      "inseecommuneprinc": "01007"
    },
    {
      "inseecommuneprinc": "01007"
    },
    {
      "inseecommuneprinc": "01006"
    },
...

whereas https://tabular-api.data.gouv.fr/api/resources/75a43900-a916-4105-adcf-fe88f88194a8/data/?inseecommuneprinc__isnotnull&inseecommuneprinc__groupby&annee__sum
yields

{
  "data": [
    {
      "inseecommuneprinc": "01001",
      "annee__sum": 58593
    },
    {
      "inseecommuneprinc": "01002",
      "annee__sum": 181913
    },
    {
      "inseecommuneprinc": "01004",
      "annee__sum": 1353962
    },
    {
      "inseecommuneprinc": "01005",
      "annee__sum": 131522
    },
...

(without duplicates)
Possible solutions:

• understand that with only one column in the groupby and no operator, the user wants a groupby
• create another syntax